Potential predictive role of Non-HDL to HDL Cholesterol Ratio (NHHR) in MASLD: focus on obese and type 2 diabetic populations

Abstract

Introduction

This cross-sectional study was conducted to examine the association between the ratio of non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol (NHHR) and metabolic dysfunction-associated steatotic liver disease (MASLD), particularly in populations with obesity and type 2 diabetes (T2D).

Methods

The analysis included 3784 participants who were 20 years and older, using data from the National Health and Nutrition Examination Survey (NHANES) 2017–2018. The prevalence of MASLD was determined using multivariable logistic regression analysis, which calculated odds ratios (ORs) and 95% confidence intervals (CIs). Conducted was an analysis employing a smooth curve fitting approach to explore the potential nonlinear association between NHHR and MASLD. Subgroup analyses were performed based on age, sex, body mass index (BMI) and T2D status to evaluate the robustness of the results, with interaction tests conducted.

Results

NHHR showed a consistently positive association with MASLD across all models. In the fully adjusted model, compared with the low NHHR group, participants in the middle and high NHHR group were associated with higher prevalence of MASLD (OR = 1.51, 95%CI = 1.25–1.83, p < 0.001, OR = 1.97, 95%CI = 1.62–2.41, p < 0.001, respectively). This positive relationship was significant across all subgroups, confirming a robust association between NHHR and MASLD.

Conclusions

This cross-sectional study found a significant linear positive relationship between NHHR and MASLD, which remained significant across different age, sex, BMI and T2D groups. These findings suggest that NHHR may have the potential to serve as a predictor for screening MASLD in populations with obesity or T2D.

Introduction

Metabolic Dysfunction-associated steatotic liver disease(MASLD) is an updated definition proposed in 2023 by an international expert group in a multisociety Delphi consensus statment, which has been used as an alternative term for non-alcoholic fatty liver disease (NAFLD), utilizing 'positive criteria' for diagnosis while serving as a more fitting descriptor for stages characterized by established metabolic dysfunction [1, 2]. Diagnostic parameters for MASLD rely on demonstrating hepatic steatosis alongside at least one of five cardiometabolic criteria [2]. Importantly, individuals with MASLD who consume greater amounts of alcohol per week are categorized as having metabolic and alcohol relate/associated liver disease (MetALD) [2]. Positive identification is based on histological (biopsy), imaging studies, or blood biomarkers indicating hepatic fat accumulation [3]. MASLD affects about a quarter of the world's adult population and imposes a significant health and economic burden on all societies [4]. However, there are currently no approved drug therapies. At the same time, the prevalence of poor metabolic health is high among adults from wealthy countries [5]. In particular, the prevalence of MASLD in the U.S. population was as high as 39.1% in the 2017–2018 National Health and Nutrition Examination Survey (NHANES) results [6].

MASLD and metabolic syndrome are closely linked [7]. It is common for patients with MASLD to also have metabolic syndrome features like obesity, diabetes, and dyslipidemia [8]. Furthermore, an escalation in the number of metabolic syndrome components appears to elevate the likelihood of developing MASLD [9]. A recent meta-analysis found that close to 70% of individuals diagnosed with type 2 diabetes mellitus (T2D) also suffer from MASLD. Clinically significant fibrosis raises the likelihood of decompensated cirrhosis, hepatocellular carcinoma, liver transplantation, and overall mortality, especially in individuals with T2D [10]. However, clinical practitioners underestimate its prevalence and inconsistently implement appropriate screening strategies, thus missing the diagnosis of MASLD in high-risk populations such as obese individuals and those with T2D. This pattern of underdiagnosis is further compounded by limited referrals to specialists and insufficient prescriptions, leading to increased complexity of the condition. Current guidelines also recommend paying attention to screening for MASLD in individuals with T2D, aiming for a comprehensive management of both conditions [11]. Therefore, exploring feasible screening methods is crucial.

MASLD is closely related to disorders of lipid metabolism and the inflammatory mechanisms they cause. Previous studies have shown that high-density lipoprotein (HDL) possesses a variety of antioxidant and anti-inflammatory properties, that impaired antioxidant capacity of HDL may influence disease severity in MASLD patients [12], and that oxidised low-density lipoprotein (LDL) and very low density lipoprotein (VLDL) have the ability to induce severe lipid oxidation and oxidative damage to DNA, and to cause hepatic injury by promoting inflammatory responses through binding to scavenger receptors on Kupffer cells [13–15].

The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio (NHHR) is a newly introduced comprehensive lipid index that has demonstrated strong predictive value for cardiovascular and cerebrovascular diseases, surpassing traditional lipid parameters [16–18]. In addition, studies have shown that NHHR independently predicts the likelihood of developing diabetes, which may be due to the fact that lipid disorders are closely related to insulin resistance and glucose metabolism disorders [19, 20]. NHHR has also been validated as an independent predictor of chronic kidney disease, which may be due to the fact that renal impairment may be associated with higher concentrations of VLDL particles and atherogenic small LDL and IDL particles, and that lipoprotein abnormalities and impaired renal function are causally related and interact with each other [21, 22]. However, there is currently no research indicating whether this predictive effect remains significant in individuals with or without T2D.

To address this gap, our research involved a cross-sectional analysis based on data from NHANES 2017 to 2018. The main objective of our research was to investigate the association between NHHR and MASLD, specifically in groups with obesity and T2D, and to determine its effectiveness in predicting MASLD in these populations.

Materials and methods

Study population

Using NHANES data, this cross-sectional study aims to determine the significance of NHHR in relation to MASLD. NHANES is an extensive survey conducted nationally to observe the health and nutritional conditions of individuals throughout the United States. What sets it apart is its integration of interviews, physical assessments, and laboratory analyses. The data collected in NHANES is categorized into demographics data, dietary data, examination data, laboratory data and questionnaire data. The National Center for Health Statistics (NCHS), a part of the Centers for Disease Control and Prevention (CDC), is responsible for managing NHANES. The NHANES survey protocol has been approved by the Ethics Review Committee of the NCHS, and participants have given written consent for the use of their data in studies.

We gathered data from 9254 participants in NHANES 2017–2018 in alignment with the study's objectives, excluding individuals with incomplete data on MASLD (n = 4135), high-density lipoprotein cholesterol (HDL-C) and total cholesterol (TC) (n = 275), baseline covariates (n = 793), and those under the age of 20 (n = 191). Extreme values of NHHR (< 1st percentile and > 99th percentile, n = 76) were also excluded to prevent potential outliers from influencing the analysis. Finally, After all, 3784 subjects were taken into account for analysis, as illustrated in Fig. 1.

Fig. 1.

Flowchart of participants enrollment. Abbreviations: NHANES: the National Health and Nutrition Examination Survey, MASLD: metabolic dysfunction-associated steatotic liver disease, HDL-C: high-density lipoprotein cholesterol, TC: total cholesterol, NHHR: the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio, PIR: poverty income ratio, BMI: body mass index, WC: waist circumference, HbA1c: glycohemoglobin, ALT: alanine aminotransferase, AST: aspartate aminotransferase

Measurement of NHHR

Data on HDL-C was sourced from the HDL file, while TC data was sourced from the TCHOL file. Both measurements were using the Roche Cobas 6000 Chemistry Analyzer at Advanced Research and Diagnostic Laboratory (ARDL). The computation of Non-HDL-C levels involved subtracting HDL-C from TC, and NHHR was determined by dividing Non-HDL-C by HDL-C. NHHR was divided into tertiles for a more nuanced analysis: T1 (lowest third) included NHHR values ranging from 0.88 to 2.10, T2 (middle third) encompassed NHHR values between 2.10 and 3.10, and T3 (highest third) consisted of NHHR values exceeding 3.10, up to 7.13.

Definition of MASLD

The proposed criteria for a positive diagnosis of MASLD are based on histological (biopsy), imaging, or blood biomarker evidence of fat accumulation in the liver (hepatic steatosis) in addition to the presence of at least one of the following five cardiometabolic criteria, including (1)BMI ≥ 25kg/m²[23 Asia] or WC > 94cm/80cm[female] or ethnicity adjusted equivalent; (2)fasting serum glucose ≥ 5.6mmol/L [100 mg/dl] or 2-h post-load glucose levels ≥ 7.8mmol/L [≥ 140 mg/dl] or HbA1c ≥ 5.7% [39mmol/L] or type 2 diabetes or treatment for type 2 diabetes; (3)blood pressure ≥ 130/85mmHg or specific antihypertensive drug treatment; (4) plasma triglycerides ≥ 1.70mmol/L [150mg/dl] or lipid lowering treatment; (5)plasma HDL-C ≤ 1.0mmol/L [40mg/dl] (male) and ≤ 1.3mmol/L [50mg/dl] (female) or lipid lowering treatment [2].

The level of hepatic steatosis was evaluated using FibroScan®, which measures ultrasound attenuation and reports the controlled attenuation parameter (CAP) as an indicator of liver fat content. CAP data was obtained from the LUX file. Hepatic steatosis was characterized by median CAP scores of ≥ 248 dB/m, based on the defined criteria set by a detailed meta-analysis [23] and consistent with prior research [24–26].

Covariates

The demographics variables comprised age, sex, race/ethnicity (classified as Non-Hispanic White, Non-Hispanic Black, Mexican American and other Race), education level (categorized as less than high school, high school and more than high school) and poverty income ratio (PIR). Smoking status were divided into three groups: never (having smoked less than 100 cigarettes in life), former (having smoked over 100 cigarettes in life but not currently smoking) and now (having smoked over 100 cigarettes and currently smoke occasionally or daily). The examination factors encompassed body mass index (BMI)and waist circumference (WC). Among the laboratory variables were glycohemoglobin (HbA1c), alanine aminotransferase (ALT), aspartate aminotransferase (AST), and aspartate aminotransferase (GGT). Hypertension was defined by an average systolic blood pressure of 140mmHg or higher, or diastolic blood pressure of 90mmHg or higher. Participants were diagnosed with T2D if they met any of the following criteria and were not diagnosed with type 1 diabetes: (1) receiving a diagnosis of diabetes from a healthcare professional, (2) HbA1c > 6.5%, (3) fasting glucose level of ≥ 7.0mmol/L, (4) random blood glucose level of ≥ 11.1mmol/L, (5) two-hour post-load glucose levels blood glucose level of ≥ 11.1mmol/L, (6) use of medication for diabetes or insulin.

Statistical analysis

Mean ± standard deviation (SD) was used to display continuous variables, while sample size and its proportion [n (%)] were used for categorical variables. Comparisons between groups based on NHHR tertiles were made using the t-test for continuous variables and the chi-squared test for categorical variables.

Multivariable logistic regression analysis was employed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) to determine the prevalence of MASLD. Three models included non-adjusted model; age, sex, race/ethnicity, education level and PIR adjusted model 1; age, sex, race/ethnicity, education level, PIR, smoking status, BMI, WC, HbA1c, ALT, AST, GGT, hypertension and T2D adjusted model 2. A trend analysis was conducted among NHHR tertile groups. A smooth curve was used to explore potential linear association between NHHR and MASLD. Subgroup analyses were carried out according to age (< 65 years and ≥ 65 years), sex (Male and Female), BMI (< 30 kg/m²and ≥ 30 kg/m²) and T2D (No and Yes) to examine the robustness of the results and interaction test was conducted.

Sample weights were not applied, as the analysis focused on data from a single cycle (2017–2018). Omitting weights provided a clearer interpretation of the results without distorting subgroup-specific effects.

Data were analyzed using the R® software package (v.4.2.0, http://www.r-project.org, accessed on 22 April 2022) and Empower® software (v.4.2, http://www.empowerstats.com, X&Y Solutions, Inc. Boston, MA, USA). Statistical significance was determined using a significance level of p < 0.05.

Results

Baseline characteristics of participants stratified by NHHR

As shown in Table 1, NHHR tertiles were used to categorize all 3784 participants into three groups: low (0.88–2.10, n = 1261), middle (2.10–3.10, n = 1261), and high (3.10–7.13, n = 1262). Age did not differ in three groups. Participants with low NHHR exhibited lower prevalence of MASLD, hypertension, and T2D compared to those with middle and high NHHR. Furthermore, individuals in the high NHHR group were more males, Mexican American and other race and current smoker compared with low and moderate NHHR populations. They have lower education level, lower values of PIR and HDL-C, and higher values of BMI, WC, HbA1c, ALT, AST, GGT and TC.

Table 1.

Baseline characteristic of participants based on NHHR stratifcation (n = 3784)

Covariates	Total	NHHR			P-value
		T1 (0.88–2.10)	T2 (2.10–3.10)	T3 (3.10–7.13)
N	3784	1261	1261	1262
Age (years), Mean ± SD	50.87 ± 17.30	51.05 ± 19.00	51.03 ± 17.14	50.52 ± 15.59	0.570
Sex, n (%)					< 0.001
Male	1858 (49.10%)	490 (38.86%)	557 (44.17%)	811 (64.26%)
Female	1926 (50.90%)	771 (61.14%)	704 (55.83%)	451 (35.74%)
Race/Ethnicity, n (%)					< 0.001
Non-Hispanic White	1384 (36.58%)	453 (35.92%)	461 (36.56%)	470 (37.24%)
Non-Hispanic Black	824 (21.78%)	343 (27.20%)	293 (23.24%)	188 (14.90%)
Mexican American	494 (13.05%)	128 (10.15%)	165 (13.08%)	201 (15.93%)
Other Race	1082 (28.59%)	337 (26.72%)	342 (27.12%)	403 (31.93%)
Education Level, n (%)					< 0.001
Less than high school	682 (18.02%)	197 (15.62%)	206 (16.34%)	279 (22.11%)
High school	907 (23.97%)	297 (23.55%)	290 (23.00%)	320 (25.36%)
More than high school	2195 (58.01%)	767 (60.82%)	765 (60.67%)	663 (52.54%)
PIR, Mean ± SD	2.58 ± 1.61	2.72 ± 1.63	2.57 ± 1.62	2.46 ± 1.59	< 0.001
Smoking status, n (%)					< 0.001
Never	2177 (57.53%)	777 (61.62%)	739 (58.60%)	661 (52.38%)
Former	938 (24.79%)	292 (23.16%)	292 (23.16%)	354 (28.05%)
Now	669 (17.68%)	192 (15.23%)	230 (18.24%)	247 (19.57%)
BMI (kg/m2), Mean ± SD	29.69 ± 7.11	27.23 ± 6.61	30.29 ± 7.22	31.53 ± 6.78	< 0.001
WC (cm), Mean ± SD	100.67 ± 16.97	93.94 ± 16.16	101.77 ± 16.88	106.31 ± 15.48	< 0.001
HbA1c, n (%)	5.85 ± 1.08	5.70 ± 0.90	5.81 ± 0.98	6.05 ± 1.29	< 0.001
ALT (U/L), Mean ± SD	22.48 ± 16.43	19.47 ± 14.57	21.32 ± 14.41	26.65 ± 19.05	< 0.001
AST (U/L), Mean ± SD	22.03 ± 12.71	22.02 ± 14.96	21.07 ± 9.79	23.01 ± 12.77	< 0.001
GGT (U/L), Mean ± SD	32.33 ± 46.19	29.45 ± 44.46	28.60 ± 37.57	38.93 ± 54.32	< 0.001
HDL-Cholesterol (mg/dL), Mean ± SD	53.00 ± 14.62	64.99 ± 14.63	52.14 ± 9.88	41.90 ± 7.89	< 0.001
Total Cholesterol (mg/dL), Mean ± SD	187.92 ± 39.43	167.53 ± 34.68	185.58 ± 33.63	210.62 ± 37.37	< 0.001
MASLD, n (%)					< 0.001
No	2421 (63.98%)	770 (61.06%)	523 (41.48%)	313 (24.80%)
Yes	1363 (36.02%)	491 (38.94%)	738 (58.52%)	949 (75.20%)
Hypertension, n (%)					< 0.001
No	2054 (54.28%)	953 (75.57%)	796 (63.12%)	672 (53.25%)
Yes	1730 (45.72%)	308 (24.43%)	465 (36.88%)	590 (46.75%)
T2D, n (%)					< 0.001
No	2987 (78.94%)	1027 (81.44%)	1007 (79.86%)	953 (75.52%)
Yes	797 (21.06%)	234 (18.56%)	254 (20.14%)	309 (24.48%)

NHHR The non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio, PIR Poverty income ratio, BMI Body mass index, WC Waist circumference, HbA1c Glycohemoglobin, ALT Alanine aminotransferase, AST Aspartate aminotransferase, GGT Gamma glutamyl transferase, MASLD Metabolic dysfunction–associated steatotic liver disease, T2D Type 2 diabetes

Association of NHHR and MASLD

In Table 2, we employed three models for adjustment: the non-adjusted Model, Model 1, and Model 2. Model 1 was controlled for age, sex, race/ethnicity, education level, and PIR, while Model 2 additionally included adjustments for smoking status, BMI, WC, HbA1c, ALT, AST, GGT, hypertension, and T2D. NHHR showed a consistently strong positive association with MASLD across all models. In the fully adjusted model (Model 2), with each unit increasing in NHHR, the odds of MASLD increased by 26% among participants (OR = 1.26, 95%CI = 1.18–1.35, p < 0.001). Compared with the low NHHR group, participants in the middle and high NHHR group were associated with higher prevalence of MASLD (OR = 1.51, 95%CI = 1.25–1.83, p < 0.001, OR = 1.97, 95%CI = 1.62–2.41, p < 0.001, respectively) and p for trend test was < 0.001. Furthermore, the association between NHHR and MASLD was shown in the smooth curve fitting (Fig. 2).

Table 2.

Association of NHHR and MASLD

	Non-adjusted Model	Model 1	Model 2
	OR (95% CI) p value	OR (95% CI) p value	OR (95% CI) p value
NHHR
Continuous	1.40 (1.32, 1.48) < 0.001	1.44 (1.35, 1.54) < 0.001	1.26 (1.18, 1.35) < 0.001
T1 (0.88–2.10)	Ref	Ref	Ref
T2 (2.10–3.10)	1.81 (1.52, 2.15) < 0.001	1.90 (1.59, 2.27) < 0.001	1.51 (1.25, 1.83) < 0.001
T3 (3.10–7.13)	2.72 (2.29, 3.22) < 0.001	2.94 (2.45, 3.53) < 0.001	1.97 (1.62, 2.41) < 0.001
p for trend	< 0.001	< 0.001	< 0.001

Non-adjusted Model: adjusted for none

Model 1: age, sex, race/ethnicity, education level and PIR were adjusted

Model 2: age, sex, race/ethnicity, education level, PIR, smoking status, BMI, WC, HbA1c, ALT, AST, GGT, hypertension and T2D were adjusted

Abbreviations: NHHR: the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio, MASLD: metabolic dysfunction–associated steatotic liver disease, PIR: poverty income ratio, BMI: body mass index, WC: waist circumference, HbA1c: Glycohemoglobin, ALT: alanine aminotransferase, AST: aspartate aminotransferase, GGT: gamma glutamyl transferase, T2D: type 2 diabetes

Fig. 2.

The association between NHHR and MASLD. Each red dot represents the NHHR score level, forming a continuous fitting curve. The area between the blue dashed lines is considered as 95% confidence interval. Age, sex, race/ethnicity, education level, PIR, smoking status, BMI, WC, HbA1c, ALT, AST, GGT, hypertension and T2D were adjusted. Abbreviations: NHHR: the non-high-density lipoprotein cholesterol to high-density lipoprotein cholesterol ratio, MASLD: metabolic dysfunction–associated steatotic liver disease, PIR: poverty income ratio, BMI: body mass index, WC: waist circumference, HbA1c: Glycohemoglobin, ALT: alanine aminotransferase, AST: aspartate aminotransferase, GGT: gamma glutamyl transferase, T2D: type 2 diabetes

Subgroup analysis

To further explore the specific population on the association between NHHR and MASLD, we conducted subgroup analyses for age, sex, BMI, and T2D (Fig. 3). The results showed that the positive relationship was significant in all the subgroups, reconfirming the existence of a robust positive association between NHHR and MASLD. Furthermore, we found that the strength of the association between NHHR and MASLD was possibly greater in age < 65 years, BMI < 30 kg/m² and non-T2D population, although only the BMI and T2D subgroup had p for interaction < 0.05.

Fig. 3.

Subgroup analysis for the association of NHHR and MASLD. Solid red dots represent sample size. The black horizontal line represents 95% confidence interval. Age, sex, race/ethnicity, education level, PIR, smoking status, BMI, WC, HbA1c, ALT, AST, GGT, hypertension and T2D were adjusted (in addition to the stratification factor itself). Abbreviations: PIR: poverty income ratio, BMI: body mass index, WC: waist circumference, HbA1c: Glycohemoglobin, ALT: alanine aminotransferase, AST: aspartate aminotransferase, GGT: gamma glutamyl transferase, T2D: type 2 diabetes

Discussion

The results of our cross-sectional analysis on 3784 U.S. adults imply the linear association between high NHHR scores and the presence of MASLD. This association was stably significant in a variety of subgroups, including age, sex, BMI, and T2D, and particularly showed a higher prevalence in non-older (age < 65 years), non-obese (BMI < 30 kg/m²), and non-T2D population, but only the BMI and T2D subgroup had p for interaction < 0.05.

Replacing NAFLD with MASLD has become an international trend, because NAFLD does not reflect the current knowledge and metabolic dysfunction associated with liver disease [3]. Previous research indicates that MASLD is a liver condition associated with obesity and metabolic syndrome, demonstrating strong links to obesity, insulin resistance, hypertension, and dyslipidemia [7–9, 27]. Several similar studies have previously been conducted on Chinese participants, suggesting that the non-HDL-C/HDL-C ratio may be a viable predictor of NASH or NAFLD [28, 29]. One longitudinal cohort study based on a Chinese non-obese population found that the low-density lipoprotein cholesterol (LDL-C)/HDL-C ratio was superior to HDL-C and LDL-C in predicting new-onset NAFLD [30]. Another study based on a US population found that elevated non-HDL-C/HDL-C levels were independently associated with an increased risk of NAFLD and liver fibrosis, again supporting our results [31]. However, given that obesity is associated with a greater chance of developing MASLD in T2D, it is particularly important to predict MASLD in these populations.

Research conducted previously has revealed that individuals suffering from MASLD showcase increased serum triglyceride levels and decreased HDL levels [32]. Meanwhile, the excessive production of VLDL is a key metabolic disorder, it induces severe lipid oxidation and DNA oxidative damage, which affects the extent of liver injury [13, 14]. Elevated VLDL levels and concurrent hypertriglyceridemia support the synthesis of LDL [33], and oxidized LDL promotes an inflammatory response by binding to scavenger receptors present on Kupffer cells [34]. However, overexpression of inflammatory cytokines inhibits cholesterol excretion via bile acids and activates cholesterol synthesis [35]. Free cholesterol is lipotoxic and can drive aseptic inflammation by interacting with YAP-TAZ, which is also significantly increased in the liver tissue of human MASLD patients and mouse livers in MASLD models [15]. It also produces inflammatory cytokines by acting on hepatic Kupffer cells and adipocytes, destroying hepatocytes and activating Kupffer cells, creating an inflammatory cycle [15]. In addition, excess endogenous cholesterol activates hepatic X receptors (LXR), which regulate cholesterol homeostasis, induce hepatic steatosis, and promote hepatic secretion of more VLDL particles [36]. And MASLD patients can effectively reduce cardiovascular disease and ameliorate liver injury with cholesterol-lowering therapy [37]. This provides a rationale for NHHR to be a predictor of MASLD.

Another noteworthy result in this study is that the results of subgroup analyses indicated that BMI and T2D were the interaction term in the relationship between NHHR and MASLD. The positive association between NHHR and MASLD persisted regardless of whether one was obese or had T2D in the BMI and T2D subgroup, which shows that NHHR is an standalone predictor of MASLD regardless of BMI and T2D. However, a higher prevalence was shown in the non-obese population, which may be due to the fact that non-obese MASLD patients have a greater amount of visceral adipose tissue than healthy controls [38]. Visceral fat plays a larger role in MASLD development than overall body fat [39]. It is also possible that there is an association with lower skeletal muscle mass in non-obese MASLD patients, as muscle loss is associated with MASLD complications, such as MASH and liver fibrosis [40]. This suggests to us that although MASLD is strongly associated with obesity, perhaps screening and testing for MASLD in non-obese populations should also be of interest. The prevalence of MASLD was slightly higher in non-T2D patients than in T2D patients in this study. A study suggests that in the absence of diabetes mellitus, although there may be no obvious effect of disorders of glucose metabolism and insulin resistance on MASLD, disorders of lipid metabolism may still play a central role in it, especially in atherosclerosis formation and development [41]. However, the exact mechanism remains unclear. Moreover, existing studies have demonstrated that insulin-resistant patients increase VLDL secretion when trying to maintain hepatic lipid homeostasis [42]. Both insulin and glucose can drive adipogenesis through their respective transcription factors [43], thereby causing hepatic steatosis. Again, more studies are still needed to validate this association due to the cross-sectional design of this study. Importantly, based on epidemiologic studies, as described in the previous part of this study, there is a significant co-morbidity between obesity or diabetes and MASLD, which is why these two groups are the ones we focus on. Our study suggests that NHHR is positively associated with MASLD in obese (BMI ≥ 30kg/m²) or in T2D population, which means that NHHR have the potential as the predictor for screening MASLD in the two specific population.

NHHR represents an innovative lipid profile marker for atherosclerosis, encompassing details on both atherogenic and anti-atherogenic lipid particles connected to dyslipidemia-related issues [20]. A growing number of studies have found NHHR to be a superior indicator of lipid-related diseases (including metabolic syndrome, MASLD, coronary heart disease, etc.) [18, 44, 45]. Kwok RM et al. also demonstrated that NHHR is a stronger predictor of MASLD than other lipid indicators [46]. Furthermore, this indicator showed higher precision in forecasting diseases linked to diabetes development compared to standard lipid screening [47]. In conclusion, the NHHR has demonstrated a high degree of superiority in a variety of studies, and the assay is noninvasive and safe, cost-effective, and readily available, offering a promising future for clinical implementation.

The strengths of our study include the following points.To our knowledge, this is the initial examination of the link between NHHR and MASLD in a substantial, diverse sample of the U.S. population. Furthermore,our subgroup analyses also supported the strong positive association between NHHR and MASLD in all subgroups including age, gender, BMI, and T2D. However, our study still has some limitations. Because of the constraints of the study design, we could not establish a causal link between NHHR and MASLD, only that the two were associated. Despite best efforts to adjust for confounding variables, the potential impact of certain relevant confounders could not be completely avoided. Nonetheless, it provides insights for further analysis or experimental studies.

Conclusion

This cross-sectional study found a significant linear positive relationship between NHHR and MASLD, which remained significant across different age, sex, BMI and T2D groups. These findings suggest that NHHR may have the potential to serve as a predictor for screening MASLD in populations with obesity or T2D population. The use of gold standard diagnostic methods, such as liver biopsy or MRI, is recommended to further substantiate this finding.

Authors’ contributions

The study was designed and the data was collected and analyzed by Xiao-Man Ma. Xiao-Man Ma,  Yu-Miao Guo, Shu-Yi Jiang and Ke-Xuan Li were responsible for writing and modifying the manuscript. Xiao-Man Ma and Yu-Miao Guo created the tables and drew the figures. Ya-Fang Zheng assisted in extracting data. Xu-Guang Guo and Zhi-Yao Ren reviewed and made modifications to the manuscript.

Funding

This research was funded by the Natural Science Foundation of China (82201954 to Z.R.).

Data availability

The NHANES data sets can be accessed by the public through the Centers for Disease Control and Prevention website at https://wwwn.cdc.gov/nchs/nhanes/Default.aspx.

Declarations

Ethics approval and consent to participate

The survey protocol for the NHANES received approval from the Institutional Research Ethics Review Board of CDC's National Center for Health Statistics. All participants gave written informed consent, and the NCHS Research Ethics Review Board also approved the study (https://wwwn.cdc.gov/nchs/nhanes/default.aspx).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.